s2io: netpoll support
Subsystems (2)
PATCH 1/1 s2io: netpoll support
This adds netpoll support for things like netconsole/kgdboe to the s2io 10GbE driver.
This duplicates some code from s2io_poll() as I wanted to be least-invasive, someone from Neterion might have other thoughts?
drivers/net/s2io.c +51 -0
@@ -2575,6 +2575,53 @@ no_rx:#endif /** +* s2io_netpoll - Rx interrupt service handler for netpoll support +* @dev : pointer to the device structure. +* Description: +* Polling 'interrupt' - used by things like netconsole to send skbs +* without having to re-enable interrupts. It's not called while +* the interrupt routine is executing. +*/ + +#ifdef CONFIG_NET_POLL_CONTROLLER +static void s2io_netpoll (struct net_device *dev) +{ +nic_t *nic = dev->priv; +mac_info_t *mac_control; +struct config_param *config; +XENA_dev_config_t __iomem *bar0 = nic->bar0; +u64 val64; +int i; + +atomic_inc (&nic->isr_cnt); + +/* Disable all interrupts */ +en_dis_able_nic_intrs (nic,ENA_ALL_INTRS ,DISABLE_INTRS ); + + mac_control = &nic->mac_control; + config = &nic->config; + + val64 =readq (&bar0->rx_traffic_int); +writeq (val64, &bar0->rx_traffic_int); + +for (i =0 ; i < config->rx_ring_num; i++) +rx_intr_handler (&mac_control->rings[i]); + +for (i =0 ; i < config->rx_ring_num; i++) { +if (fill_rx_buffers (nic, i) == -ENOMEM ) { +DBG_PRINT (ERR_DBG ,"%s:Out of memory" , dev->name); +DBG_PRINT (ERR_DBG ," in Rx Netpoll!!\n" ); +break ; + } + } +/* Re-enable all interrupts */ +en_dis_able_nic_intrs (nic,ENA_ALL_INTRS ,ENABLE_INTRS ); +atomic_dec (&nic->isr_cnt); +return ; +} +#endif + +/** * rx_intr_handler - Rx interrupt handler * @nic: device private variable. * Description:
@@ -6210,6 +6257,10 @@ Defaulting to INTA\n"); dev->weight =32 ;#endif +#ifdef CONFIG_NET_POLL_CONTROLLER + dev->poll_controller = s2io_netpoll; +#endif + dev->features |=NETIF_F_SG |NETIF_F_IP_CSUM ;if (sp->high_dma_flag == TRUE) dev->features |=NETIF_F_HIGHDMA ;
I don't think we should disable and enable all interrupts in the poll_controller entry point. With the current patch, at the end of the routine _all_ interrupts get enabled which is not desirable. Maybe you should just do disable_irq() at start of function and enable_irq() before exiting, the way some of the other drivers do.
Ravi
-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of Brian Haley
Sent: Thursday, June 08, 2006 9:02 AM
To: [email protected]
Cc: [email protected]
Subject: [PATCH] s2io: netpoll support
This adds netpoll support for things like netconsole/kgdboe to the s2io 10GbE driver.
This duplicates some code from s2io_poll() as I wanted to be least-invasive, someone from Neterion might have other thoughts?
This adds netpoll support for things like netconsole/kgdboe to the s2io 10GbE driver.
drivers/net/s2io.c +48 -0
@@ -2575,6 +2575,50 @@ no_rx:#endif /** +* s2io_netpoll - Rx interrupt service handler for netpoll support +* @dev : pointer to the device structure. +* Description: +* Polling 'interrupt' - used by things like netconsole to send skbs +* without having to re-enable interrupts. It's not called while +* the interrupt routine is executing. +*/ + +#ifdef CONFIG_NET_POLL_CONTROLLER +static void s2io_netpoll (struct net_device *dev) +{ +nic_t *nic = dev->priv; +mac_info_t *mac_control; +struct config_param *config; +XENA_dev_config_t __iomem *bar0 = nic->bar0; +u64 val64; +int i; + +disable_irq (dev->irq); + +atomic_inc (&nic->isr_cnt); + mac_control = &nic->mac_control; + config = &nic->config; + + val64 =readq (&bar0->rx_traffic_int); +writeq (val64, &bar0->rx_traffic_int); + +for (i =0 ; i < config->rx_ring_num; i++) +rx_intr_handler (&mac_control->rings[i]); + +for (i =0 ; i < config->rx_ring_num; i++) { +if (fill_rx_buffers (nic, i) == -ENOMEM ) { +DBG_PRINT (ERR_DBG ,"%s:Out of memory" , dev->name); +DBG_PRINT (ERR_DBG ," in Rx Netpoll!!\n" ); +break ; + } + } +atomic_dec (&nic->isr_cnt); +enable_irq (dev->irq); +return ; +} +#endif + +/** * rx_intr_handler - Rx interrupt handler * @nic: device private variable. * Description:
@@ -6210,6 +6254,10 @@ Defaulting to INTA\n"); dev->weight =32 ;#endif +#ifdef CONFIG_NET_POLL_CONTROLLER + dev->poll_controller = s2io_netpoll; +#endif + dev->features |=NETIF_F_SG |NETIF_F_IP_CSUM ;if (sp->high_dma_flag == TRUE) dev->features |=NETIF_F_HIGHDMA ;
Since the poll_controller entry point will be used by utilities such as netdump, I am thinking we need to clear Tx interrupts as well here. Did you get a chance to test this patch with netdump ?
Thanks,
Ravi
-----Original Message-----
From: Brian Haley [mailto:[email protected]]
Sent: Thursday, June 15, 2006 11:37 AM
To: [email protected]
Cc: [email protected]; [email protected]; Ananda. Raju (E-mail); Leonid. Grossman (E-mail)
Subject: Re: [PATCH] s2io: netpoll support
This adds netpoll support for things like netconsole/kgdboe to the s2io 10GbE driver.
Quoted 3 lines · Ravinandan Arakali
Since the
poll_controllerentry point will be used by utilities such as netdump, I am thinking we need to clear Tx interrupts as well here. Did you get a chance to test this patch with netdump ?
No, I've only been testing Kgdb over Ethernet on Debian, I think netdump is a Red Hat thing. I can test any changes you'd want to make with this patch if you post the diffs.
-Brian
Quoted 18 lines
-----Original Message-----
From: Brian Haley [mailto:[email protected]]
Sent: Thursday, June 15, 2006 11:37 AM
To: [email protected]
Cc: [email protected]; [email protected]; Ananda. Raju (E-mail); Leonid. Grossman (E-mail)
Subject: Re: [PATCH] s2io: netpoll supportThis adds netpoll support for things like netconsole/kgdboe to the s2io 10GbE driver.
Signed-off-by Brian Haley <[email protected]>-
To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html